INDEX

Explanations

instances of objects or individuals being visually detected or observed

oai_token-act-pair · gpt-3.5-turbo

instances of the word "spotted"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

externalActionCode

-0.91

industrial

-0.81

mort

-0.80

vernment

-0.80

nesia

-0.79

cknow

-0.79

venge

-0.76

idem

-0.76

bern

-0.76

anwhile

-0.75

POSITIVE LOGITS

 spotted

1.13

bugs

0.96

owl

0.89

 spotting

0.85

Owl

0.83

 Viper

0.75

 rods

0.75

 sightings

0.74

 spots

0.73

swick

0.70

Activations Density 0.012%

instances of objects or individuals being visually detected or observed

instances of the word "spotted"

No Comments

No Known Activations